SMUT

SMUT tests mediation by multiple correlated single nucleotide polymorphisms (SNPs) on phenotypic outcomes through mediators such as gene expression in Genome-Wide Association Studies (GWAS).

Key Features:

• Simultaneous Testing of Multiple Correlated SNPs: Tests mediation effects of multiple correlated SNPs through a mediator (for example, nearby gene expression) on phenotypic outcomes.
• Intersection-Union Test Framework: Employs an intersection-union test to evaluate mediation hypotheses across multiple SNPs.
• Powerful Simulation Validation: Extensive simulations demonstrated validity with reported power gains up to 92% versus alternative methods.
• Real-World Application and Validation: Confirmed known mediators in real datasets, including plasma adiponectin levels in Finnish populations, where many alternative methods did not detect them.

Scientific Applications:

• Mechanistic hypothesis generation: Facilitates generation of hypotheses linking non-coding GWAS variants to phenotypes via mediators such as gene expression.
• Functional follow-up prioritization: Prioritizes candidate mediators for experimental validation in functional follow-up studies.
• Polygenic trait analysis: Suited for studying polygenic traits by accommodating multiple correlated SNPs that jointly influence mediators and phenotypes.

Methodology:

Integrates mediation analysis with an intersection-union test framework to perform simultaneous testing of multiple correlated SNPs through mediators such as gene expression.